# dots on maps

[Jim Croft]

> My problems with vouchers/non-vouchers are based partly on Neil's comments
> about shot-guns but more importantly on several other thoughts.

There is no problem using both vouchered data and unvouchered survey
data, as long as it is totally unambiguous what you are dealing with.
Used together they can paint a very powerful picture.  And each class of
data can be used check, and improve the quality of the other.  I guess
that is what maps are: a falsifiable hypothesis that needs testing (or
ground-truthing). And that is what science is about, right?

> As to scalability of gridded data it depends upon the grid size and the
> accuracy of the underlying data.  Much of the older specimens we use have
> very vague location data (anywhere within probably 2-4 km of a village).  By
> putting a dot on the map you suggest that you have more detailed data.
> Obviously the data we are collected now is accurate to within a few hundred
> meters.

This too is not a problem, just as long as you know and record what the
accuracy and precision (and preferebly how you worked out this value)
of the records are.  We all have to deal with specimens that are know
to only region state, province or even country level.  They still be
used put a blob on a map, only it is a very big and very fuzzy blob.
Once you know what the precison of data are, they can be filtered in
and out depending on the question at hand and the scale and resolution
of the answer required.

just love this data talk...

jim

__________________________________________________________________________
Jim Croft ~ jrc at anbg.gov.au ~ http://www.anbg.gov.au/people/croft.jim.html
ph 02-6246-5500 ~ fx 02-6246-5248 ~ GPO Box 1600 Canberra ACT 2601